Early Cancer Detection and Diagnosis using Deep Learning Models in Medical Imaging

Breast cancer is a leading cause of mortality among women globally. While its exact cause remains uncertain, no definitive methods exist for its prevention or treatment. However, early detection through mammography plays a crucial role in improving management strategies and increasing the likelihood of recovery. Mammography aids not only in early cancer detection but also in identifying various conditions and distinguishing between cancer types—benign, malignant, or normal. This study explores an advanced machine learning and image processing framework for detecting and classifying breast cancer. The approach incorporates machine learning techniques alongside feature extraction, selection, and image preprocessing to classify breast cancer and related skin diseases effectively. Key advancements include the use of a geometric mean filter to enhance image quality and AlexNet for feature extraction. The relief algorithm is employed for feature selection, and classification is performed using machine learning models such as random forest, K-nearest neighbors (KNN), least square support vector machine (LS-SVM), and Naïve Bayes. This innovative methodology provides an accurate and efficient solution for breast cancer detection through advanced image analysis techniques.